Project description
|
◦ 30 speakers were recorded five times with time gap initial recording,
one day later, one week later, one month later, two months later - in driving
environment to be used for speaker verification. One speaker read an identical
set of 5 sentences containing all the Korean phonemes 6 times and 76 ~ 77
phonetically balanced words and 144 sequences of 4 digits one time each
◦ A total of 30 speakers (Male: 15, Female: 15)
◦ Car at 2,500 cc
◦ Driving environment: 30-60km/h, fair weather, asphalt, windows closed,
audio off
◦ 7 microphones + 1 handsfree (simultaneous recording)
◦ Sampling and data format
- microphone: 24,000 Hz, 16 bit Windows wave format
- handsfree: 8,000 Hz, 16 bit Windows wave format
◦ Prompts
- 452 phonetically balanced words
- 5 sentences containing all the Korean phonemes (for all the speakers)
- 864 Quadruple digits
◦ Per speaker
- 5 sentences (6 times)
- 76 -77 phonetically balanced words
- 144 quadruple digits
◦ Recording sessions (time gap)
- 5 sessions: 1st recording, 1 day after 1st recording, 1 week after 1st
recording, 1 month after 1st recording, 2 months after 1st recording |